<![CDATA[
#!perl -w
use strict;
=head1 NAME
testio.pl
=head1 DESCRIPTION
test io impact of reading lines (sip), reading in an array (array),
or reading in the entire file (slurp).
The array reads the entire file into memory into an array. The slurp
reads the entire file into memory in a long string. The sip reads a
line at a time and writes it back.
Bio applications often need to search for patterns in one continuous
string per sequence.
=head1 SYNOPSIS
testio.pl
=head1 NOTES
Create bigfile.txt to create a benchmark. It must be large enough
to be measured in seconds.
On a slow, slow PC the timing of a 21 MB file (21,000,000 characters)
was:
Sip time 4 seconds
Array time 11 seconds
Slurp time 3 seconds
=head1 AUTHOR
David Scott from ideas on Steve Litt's Perls of Wisdom
at www.troubleshooters.com
=cut
my $bigfileName = "bigfile.txt";
my $sipfileName = "sip.out";
my $arrayfileName = "array.out";
my $slurpfileName = "slurp.out";
package main;
{
my $start = time();
print "Starting sip\n";
sip();
print "End sip\n";
my $end_sip = time();
print "Starting array\n";
buildarray();
print "End array\n";
my $end_array = time();
print "Starting slurp\n";
slurp();
print "End slurp\n";
my $end_slurp = time();
print "Sip time is ", $end_sip-$start, " seconds\n";
print "Array time is ", $end_array-$end_sip, " seconds\n";
print "Slurp time is ", $end_slurp-$end_array, " seconds\n";
}
# functions
# read and write file in one fell slurp
sub slurp {
my $in_file;
my $out_file;
local $/;
open ($in_file, "< $bigfileName")
or die("Error reading $bigfileName for benchmark: $!");
open $out_file, ">" . $slurpfileName;
my $buf = <$in_file>;
print $out_file $buf;
close $out_file;
close $in_file;
}
# read and write file one line at a time
sub sip {
my $in_file;
my $out_file;
open($in_file, "< $bigfileName")
or die("Create $bigfileName for benchmark");
open $out_file, ">" . $sipfileName;
while(<$in_file>) {
my $line = $_;
# chomp $line;
# print $out_file $line, "\n";
print $out_file $line;
}
close $out_file;
close $in_file;
}
# read and write the file as an array of lines
sub buildarray {
my $in_file;
my $out_file;
my @array;
open($in_file, "< $bigfileName")
or die("Create $bigfileName for benchmark");
open $out_file, ">" . $arrayfileName;
@array = <$in_file>;
print $out_file @array;
close $out_file;
close $in_file;
}
]]>